International Scholarly Research Network ISRN Molecular Biology Volume 2012, Article ID 691341, 10 pages doi:10.5402/2012/691341

Review Article Molecular Determinants of Cav1.2 Calcium Channel Inactivation Nikolai M. Soldatov Humgenex Inc., Kensington, MD 20895, USA Correspondence should be addressed to Nikolai M. Soldatov, [email protected] Received 24 August 2012; Accepted 13 September 2012 Academic Editors: M. W. Berchtold, E. Caffarelli, and H.-C. Lee Copyright © 2012 Nikolai M. Soldatov. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Voltage-gated L-type Cav 1.2 calcium channels couple membrane depolarization to transient increase in cytoplasmic free Ca2+ concentration that initiates a number of essential cellular functions including cardiac and vascular muscle contraction, gene expression, neuronal plasticity, and exocytosis. Inactivation or spontaneous termination of the calcium current through Cav 1.2 is a critical step in regulation of these processes. The pathophysiological significance of this process is manifested in hypertension, heart failure, arrhythmia, and a number of other diseases where acceleration of the calcium current decay should present a benefit function. The central issue of this paper is the inactivation of the Cav 1.2 calcium channel mediated by multiple determinants.

1. Introduction The voltage-gated inward Ca2+ current (ICa ) is a common mechanism of transient increase in the cytoplasmic free Ca2+ concentration triggered by cell depolarization. This form of Ca2+ signaling activates essential cellular processes including cardiac contraction [1], regulation of a smooth muscle tone [2], gene expression [3], synaptic plasticity [4] and exocytosis [5]. Complete and rapid termination of Ca2+ influx is mediated by an intricate mechanism of spontaneous calcium channel inactivation, which is crucial for preventing Ca2+ overloading of the cell during action potentials and restoration of the resting sub-μM cytoplasmic free Ca2+ concentration [6]. This paper will focus on the molecular basis and multiple determinants of the Cav 1.2 calcium channel inactivation.

2. Cav 1.2: Challenges and Solutions 2.1. Molecular Complexity. The Cav 1.2 calcium channel is an oligomeric complex composed of the α1C , α2 δ, and β subunits [7, 8]. The ion channel pore is formed by the α1C peptide (Figure 1) that is encoded by the CACNA1C gene. The auxiliary β and α2 δ subunits are essential for the functional expression and plasma membrane (PM) targeting of the channel [9, 10]. They exist in multiple genomic isoforms generated by four CACNB genes (CACNB1–4)

and three CACNA2D genes (CACNA2D1–3). All three subunits are subject to alternative splicing. Adding to the complexity of the Cav 1.2 molecular organization, β subunits tend to oligomerize [11]. All together, genomic variability, alternative splicing, and hetero-oligomerization generate a plethora of Cav 1.2 splice variants that are expressed in cells in species-, tissue-, and developmental-dependent manner, while the change of their fine balance may have significant pathophysiological consequences [12, 13]. 2.2. Challenges in the Selection of the Host Cell. Naturally occurring diversity of Cav 1.2 complicates the interpretation of data obtained from native cells, let alone the single channel data. This underlies the importance of Cav 1.2 research in recombinant expression systems where the molecular composition of the channel and the structure of its constituents are predefined. However, this experimental approach encountered the major problem of the selection of an appropriate host cell. Most of the studies of calcium channels were carried out using HEK293 cells. These cells provide high expression efficiency of recombinant Ca2+ channels but, unfortunately, contain endogenous calcium channels exhibiting Ca2+ currents up to 3 pA/pF [15, 16]. Thus, HEK293 cells allow for the adequate study of recombinant Ca2+ channels only when the amplitude of the current is large enough to ignore the contribution of the endogenous channels. Correct

2

ISRN Molecular Biology I

II

III

8/8A 7

5 + 3 + + + 12 34 5 4

H2 N 1/1A

2

13

12

6

6 9 AID 9A

14 15

+ + + + 12 34 5 11 10

IV

21/22 26 24

33 28

+ 20 + + 2 + 27 6 1 34 5 6 16

23 19

17

18

25

30

34A 36

Out

+ 37 Plasma + + membrane 35 + 12 34 5 6 38 In 34 39 29 41A 41 42 31/32 43 40 44 LA IQ HOOC 45 46 50 47 48 49

Figure 1: Transmembrane topology of the α1C subunit. To illustrate the sites of molecular diversity, the polypeptide sequence is schematically segmented according to the CACNA1C genomic map [14] and the corresponding invariant (black) and alternative (blue) exons are outlined by black bars and numbered (1–50). Four regions of homology (I–IV), each composed of 6 transmembrane segments (numbered), are believed to be folded around the central pore. α-interaction domain (AID) of a constitutive β-binding site is shown in green. LA and IQ motifs (red) constitute calmodulin-binding domain (CBD).

assessment of the functional determinants of Ca2+ channels, however, requires the use of host cells that are completely free of endogenous Ca2+ channel subunits. COS1 or COS7 cells suit this requirement well because they generate no appreciable calcium current, do not contain endogenous Ca2+ channel subunits or their precursors, and show no induction of endogenous Cav 1.2 subunits in response to the expression of the recombinant ones [17, 18]. Kinetics parameters and voltage dependence of activation and inactivation of the Cav 1.2 channel currents measured in COS1 cells are consistent with data obtained in other expression systems [19]. An important advantage of COS cells is their relatively slow division rate that allows for better control over efficiency of expression and assembly of the Cav 1.2 channel subunits of different size. 2.3. Problems of Fluorescent Labeling and Measurement. Fusion of GFP-like fluorophores to the N- and/or C-termini of the recombinant α1C or to the N-terminus of β does not markedly change the electrophysiological properties of the expressed channels, enables the application of fluorescent and FRET (fluorescent resonance energy transfer) microscopy to the study of subcellular distribution and assembly of Cav 1.2 as well as intricate aspects of molecular architecture and dynamics of the channel. The channel retains major electrophysiological characteristics unchanged when the α1C C-terminal sequence encoded by distal exons 46–50 (Figure 1, residues 1833–2138 in α1C,77 ) is replaced by ECFP. However, α1C fused by its N- or/and C-termini to EYFP is highly sensitive to photobleaching that irreversibly inactivates it. Known as fluorophore-assisted light inactivation (FALI), this interesting property limits the applicability of acceptor photobleaching for the measurements of FRET in Cav 1.2 because of uncertainty in the functional state of the channel [20]. However, the ratiometric analysis of corrected FRET between the fluorophores, fused to the tails of the α1C and/or β subunits, reflects the reversible statedependent structural rearrangements of the channel induced by the changes of transmembrane voltage under patch clamp [19, 21].

2.4. Recombinant Cav 1.2: What Does It Need for Functional Expression and How Does It Appear? Typical properties of a “wild-type” recombinant Cav 1.2 are illustrated in Figure 2(A) using an example of the ubiquitous human α1C,77 isoform (GenBank no. z34815). When the EYFP-labeled α1C was expressed in COS1 cells alone, the fluorescenttagged channel protein was diffusely distributed over the cytoplasm and did not generate measurable calcium current (Figure 2(A), panel a). The quantitative analysis of distribution of α1C between PM and the cytoplasm (Figure 2(B)) [18] confirmed lack of significant PM targeting by α1C independently on the presence of α2 δ (bars a and b). Expression of Cav β in the absence of α2 δ stimulated PM targeting of α1C , but the channel remained silent (Figure 2(A), panel c) unless α2 δ was coexpressed (panel d). Thus, β and α2 δ subunits are sufficient for the functional channel; under these experimental conditions, β subunits stimulate PM targeting of the channel complex and, in the presence of α2 δ, facilitate voltage gating of the Cav 1.2 channel. The shape and appearance of the peak calcium current shown in Figure 2(A) (panel d) is quite typical for the β2 -modulated Cav 1.2 [22]. Its major features include the relatively slow rate of ICa decay and a large fraction of the sustained ICa remaining the end of the depolarizing pulse [18]. It is clear that during long-lasting action potentials such properties may lead to pathogenic calcium overload of the cell if it is not balanced by robust compensatory mechanisms. It was the ultimate role of Cav 1.2 in defining the duration of the action potential in cardiac cells that triggered the research and development of calcium channel blockers, a class of drugs that by now has a billion dollar market. It is this role of Cav 1.2 that stimulates the current interest to the identification of molecular determinants of Cav 1.2 inactivation in hopes of finding more specific and more effective drugs. 2.5. Last but Not Least a Complication: Cav 1.2 Clustering. A single ventricular myocyte contains ∼300,000 Cav 1.2 channels, but only ∼3% of the channels are open at peak ICa [23]. Contrary to the popular belief, Cav 1.2 channels

ISRN Molecular Biology

3 α1C

(c)

Ratio PM/cytoplasm

(b)

∗

2

(a)

α1C /α2 δ

α1C /β2d

1 0.5 0 a

(d)

∗

1.5

α1C /α2 δ/β2d

200 pA 200 ms (A)

b

c

d

(a) α1C (b) α1C /α2 δ (c) α 1C /β2d (d) α1C /α2 δ/β2d

(B)

Figure 2: Role of the Cav 1.2 auxiliary subunits. (A) Epifluorescent images of the expressing COS1 cells showing distribution of EYFPN -α1C obtained with the YFP filter (scaling bars, 4 μm) and traces of the maximum calcium current recorded in response to 600 ms steps to +30 mV from the holding potential Vh = −90 mV (left). (B) Relative distribution of EYFPN -α1C in the plasma membrane (PM) over the cytoplasm in the absence (a) or presence of α2 δ (b), β2d (c), or α2 δ + β2d (d). The ratio of fluorescence intensity in PM over the area underneath PM was averaged after background subtraction in each cell. The ratio less than 1.0 indicates lack of significant PM targeting by α1C . ∗ P < 0.05 [18].

are not evenly distributed over the plasma membrane. In native neuronal [24–26] and cardiac muscle cells [27– 29] they form large clusters. Single-molecule imaging of the functional recombinant EYFPN -α1C /β2a /α2 δ channels expressed in HEK293 cells revealed clusters composed of ∼40 channels that were mobile in the plasma membrane [30]. Both the fluorescence correlation spectroscopy and fluorescence recovery after photobleaching experiments yielded a lateral diffusion constant of Dlat ≈ 0.1 μm2 /s. The functional significance of the Cav 1.2 clusters mobility is not clear. It is believed that in cardiac muscle cells such mobility may be restrained by interactions with other proteins, for example, ryanodine receptors [27]. The size of Cav 1.2 clusters and their specific density in the plasma membrane depend on the type of β subunit expressed [31]. The distance between the termini of neighbor α1C subunits varies from 67 A˚ with neuronal/cardiac β1b to 79 A˚ with vascular β3 . The highest density of Cav 1.2 clusters in the plasma membrane and the smallest cluster size were observed with β1b present. Insight into molecular mechanisms defining the architecture and properties of Cav 1.2 clusters is important for better understanding of pathophysiology of the coupling between the Cav 1.2 activity and the induced responses in Ca2+ signal transduction.

3. Voltage- and Ca2+ -Dependent Inactivation of the Cav 1.2 Calcium Channel In the case of Cav 1.2 calcium channels, two different mechanisms are in control of Ca2+ current inactivation. One mechanism is driven by Ca2+ ions on the cytoplasmic side of the plasma membrane, whereas the other depends on transmembrane voltage. Experimentally, replacement of Ca2+ for Ba2+ as the charge carrier eliminates Ca2+ -dependent inactivation (CDI) [32] so that the Ba2+ -conducting calcium channels inactivate in a voltage-dependent manner by fast

(FI) and slow (SI) mechanisms [33]. These three mechanisms of inactivation, FI, SI and CDI, and their major determinants are illustrated on Figure 3. 3.1. Ca2+ -Dependent Inactivation and Calmodulin-Binding Domain of α1C . There are several different determinants of CDI, but it was not until 1997 that the Ca2+ -sensing site of CDI had been narrowed down to a stretch of the 80-aminoacid C-terminal sequence of α1C encoded by exons 40–42 [34] (Figure 2) marked by red block in Figure 3(A) (panel a). A naturally occurring splice variation in this region in α1C,86 (Figure 3(B)) completely inhibited CDI as it is evident from the lack of deceleration of the current with Ba2+ as the charge carrier (panel c, black trace) as compared with ICa (red trace). Another characteristic feature of the inhibited CDI was lack of the current size dependence of ICa on voltage (Figure 3(B), panel d, open symbols) that stays in contrast to the U-shape dependence of the time constant of fast inactivation (τ f ) on membrane potential in the wild-type Cav 1.2 (see Figure 3(A), panel d). Two distinct sequences, L and K, were identified within this 80-amino-acid stretch whose α1C,86 -like mutations in the wild-type α1C conform to the same characteristic features [35], suggesting the existence of two adjacent CDI sensors. One of them was outlined in the K region as the calmodulin- (CaM-) binding IQ motif [36] and, later on, the link of the IQ motif to CDI as the functional Ca2+ -CaM binding site was confirmed in three independent studies [37–39] by the use of CaM mutants lacking affinity to Ca2+ . Correspondingly, the LA motif was linked to CDI as apo-CaM binding site [40–42] endowed by the resting state of the channel. A single CaM molecule tethered to this Ca2+ dependent CaM-binding domain (CBD) of α1C is the major Ca2+ sensor of the channel [43, 44]. Splice variation of α1C in CBD region of α1C,86 not only completely inhibits CDI, but also removes SI (Figure 3(B), panel c) and deprives the channel of differential sensitivity to

4

ISRN Molecular Biology α1C,86

α1C,77 I

(a)

II

L S N

III

IV

I

II

III IV

I

II

III IV

A

V

I

L

A

V

I

L

T

T

T

V

I

M

S

V

I

M

I

V

I

M

N

β

N

β

C (kDa)

(kDa)

(b)

α1C,77

200 66

+ − Ca2+

66

β1a

β

C

C (kDa)

α1C,86

200

α1C,IS-IV

200 66

β1a

β1a

Ca2+

Ca2+

(c)

α1C,IS-IV

Ca2+ Ba2+

0

60

−0.5

40 20 −1

−1 −40

0

40 (mV)

80

−1 −40

0 40 (mV)

β1a

β1a

200 ms

80

−40

Normalized I Ca τ f (ms)

Normalized I Ca τ f (ms) (e)

40 20 0

−0.5

τ f (ms)

20

200 ms

0 80

Normalized I Ca

40

−0.5

Normalized I Ca

0 60 τ f (ms)

Normalized I Ca

0

τ f (ms)

(d)

Ba2+

200 ms

200 ms

Ba2+

0

40 (mV)

80

Normalized I Ca τ f (ms) β1a

β2a

200 ms

β2a

200 ms

β2a CDI:

Yes

No

No

SI:

Yes

Missing

Inhibited

DβM:

Yes

No

No

(B)

(C)

(A)

Figure 3: Molecular determinants of Cav 1.2 inactivation. Comparison of the wild-type Cav 1.2 (A) with the same channel deprived of CDI (B) and SI (C) determinants. The five horizontal panels show (a) arrangement of critical determinants of inactivation. ADSI is composed of conserved hydrophobic amino acids in a -2 position of S6 segments in repeats II, III, and IV (yellow circles: Ala, Val, and Ile, resp.) as well as Ser residue in -1 position of IS6 (cyan circle). The CaM-binding domain (CBD) of the α1C C-tail is shown by a red rounded rectangle. A β subunit (green) binds to the α-interaction domain in the linker between repeats I and II, and, in a Ca2+ -dependent manner, to the IQ-region of the α1C subunit C-tail ([45], not shown). The distal structure of β2 (β2 CED, blue ball) binds to the CBD [46]. (b) Evidence of coimmunoprecipitation of the indicated subunits. (c) Normalized traces of IBa (black) and ICa (red), and (d) voltage dependence of ICa (red) and time constant of FI (τ f , black) are presented to illustrate CDI in (A) and lack of CDI in (B) and (C). (e) Link between CDI and differential β-subunit modulation (DβM) of Cav 1.2. (A) Differential modulation of the IBa inactivation by β1a (black trace) and β2a (green trace) in the WT Cav 1.2. Disruption of CBD (α1C,86 ) eliminates CDI and SI targeted by CDI and DβM (B). Mutation of ADSI (α1C,IS-IV ) removed CDI and fully inhibited SI so that the channel remains conducting for the duration of the depolarization stimulus (C).

ISRN Molecular Biology

5 α1C -PHC /β1a /α2 δ (a)

(b)

Ca2+ Ba2+

200 pA 200 ms

(A) PHN -α1C /α2 δ (a)

(b) Ca2+ Ba2+

200 pA 200 ms

(B) ΔN -α1C /α2 δ (a)

(b) Ca2+

100 pA 200 ms

(C)

Figure 4: Differential role of the carboxyl- and amino-terminal tails of α1C in Cav 1.2 inactivation. Shown are (a) epifluorescent images illustrating the plasma membrane targeting of the EYFP-labeled α1C (scaling bars, 4 μm) and (b) superimposed traces of the maximum IBa (black) and ICa (red) scaled to the same amplitude for α1C -PHC /β1a /α2 δ [21, 47], (A) PHN -α1C /α2 δ (B) [47], and ΔN -α1C /α2 δ (C) [47].

β-subunit modulation (Figure 3(B), panel e) in spite of the fact that β remains associated with α1C,86 (Figure 3(B), panel b). This indicates that all three properties of the channel— CDI, SI, and β-subunit modulation—are linked together [48, 49]. 3.2. Slow Inactivation. A number of evidences have been presented that amino acids confined to the distal part of S6 segments in α1C play important role in SI [50–52]. Systematic study of this region [53] outlined the “annual determinant of slow inactivation” (ADSI) as a structure composed of four highly conserved amino acids of four transmembrane segments S6, constituting the cytoplasmic end of the pore (Figure 3(A), panel a). Their simultaneous mutation (S405I in IS6, A752T in IIS6, V1165T in IIIS6, and I1475T in IVS6) generates the α1C,IS-IV channel. Analysis of the current kinetics of the α1C,IS-IV channel showed tremendous acceleration of the rapidly inactivating component (τ f ≤ 10 ms) that comprises about 50% of the total IBa (or ICa ) amplitude. Slow voltage-dependent inactivation of α1C,IS-IV is fully inhibited, and the channel remains conducting for the duration of depolarization. Replacement of Ca2+ for Ba2+ as the charge carrier (panel c) did not change significantly this pattern of inactivation, while the analysis of voltage dependence of τ f for the inactivating component of ICa through the α1C,IS-IV channel (panel d) confirmed lack of CDI. The replacement of β1a for β2a (panel e) did not change inactivation of the α1C,IS-IV channel current suggesting lack of differential β-subunit modulation, while the co-immunoprecipitation analysis (panel b) provided direct evidence of association between α1C,IS-IV and β. Taken together, results presented in Figure 3 suggest that there is a cross-talk between ADSI, CBD and β, supported by direct interactions between them and/or specific conformational folding of the constituents of the polypeptide bundle

underlying the pore. Indeed, both the interaction of β with CBD and the importance of functional conformation were directly demonstrated in live cells expressing recombinant Cav 1.2. 3.3. Role of the α1C C-tail Folding. Quantitative voltagedependent FRET microscopy combined with patch clamp in the live cell showed that the α1C subunit C-terminal tail is subject to reversible voltage-gated conformational rearrangements [21, 47]. The anchoring of the α1C C-tail to the inner leaflet of the plasma membrane via the pleckstrin homology (PH) domain fused to the C-terminus of α1C (α1C -PHC ) abolished this conformational rearrangement and inhibited both SI and CDI (Figure 4(A)) in a manner very similar to that observed with α1C,IS-IV (Figure 3(C)). This modification limiting the mobility of the α1C carboxyl terminus had major implication on Ca2+ signal transduction. CREB-dependent transcriptional activation associated with the activity of Cav 1.2 was completely suppressed in spite of robust ICa generated by the “C-anchored” channel in response to depolarization. Release of the α1C C-tail by activation of PIP2 hydrolysis upon activation of phospholipase C fully restores all these deficient functions, including SI, CDI, and the effective coupling of ICa to the CREB-dependent transcription [21]. Thus, it is specific functional folding of the α1C C-terminal tail that is crucial for inactivation. It is crucial for signal transduction because it is designed to cage the permeating Ca2+ in CaM attached to CBD and to effectively move this caged Ca2+ to downstream signaling targets associated with CREB-dependent transcription or cardiac muscle contraction [54]. Above all, this function occurs in tight coordination with extracellular stimuli activating the channel. In terms of signal transduction, SI is a lock on the inside of the channel that is released by the permeating Ca2+

6 to accelerate its closure and initiate the movement of the Cterminal tail [49]. 3.4. Role of the α1C N-Terminus. All the functions mentioned above depend also on the integrity of the α1C N-terminus. Inactivation properties of the recombinant α1C /β/α2 δ channel are not greatly altered by structural changes of the proximal part of the α1C N-tail, for example, by the fusion of a fluorescent protein [19, 21], by PH domain [47], or by alternative splicing of exons 1/1A generating the long isoform of α1C [55]. The very first functional analysis of the effect of partial deletion of the α1C N-terminus showed [56] that it is involved in inactivation while β prevents inhibition of the channel by the N-tail. Using FRET microscopy combined with patch clamp, we found that inactivation causes strong mutual reorientation of the α1C and β1a NH2 termini, but their distance vis-`a-vis the plasma membrane is not appreciably changed [19]. This relative lack of mobility is conferred by β in a manner that facilitates the channel response to voltage gating. Experiments on uncoupling of the α1C subunit N-terminal tail from the regulation of the channel were carried out in the absence of β. Anchoring of the α1C N-tail in the inner leaflet of the plasma membrane via attached PH domain created conditions when PHN α1C and α2 δ were sufficient to generate a robust inward current (Figure 4(B)). This channel, however, is deprived of CDI and any voltage-dependent inactivation. Indeed, neither Ba2+ nor Ca2+ current has shown appreciable decay (see overlapped traces). Release of the α1C N-tail upon PIP2 hydrolysis by activation of phospholipase C completely inhibited the β-deficient channel [47]. Similar properties, except a much slower activation of the current, were observed on deletion of the entire (but 4 amino acids) N-terminal tail of α1C (Figure 4(C)). With either type of uncoupling of the α1C N-terminal tail—whether through a deletion or by PM anchoring,—a delay in the activation of the whole-cell current appears to be associated with prolongation of the first latency. Single channel recordings revealed that deletion of the N-tail essentially stabilized the open state of the ΔN -α1C /α2 δ channel, which showed longer openings during long-lasting depolarization [47]. Thus, CDI is mediated by CBD determinants of the α1C C-tail, by the ADSI in the cytoplasmic pore region, and by the folding of the α1C C- and N-termini. Calmodulin integrates these determinants, providing a Ca2+ -dependent switch that terminates slow inactivation, releases the α1C Ctail, and shuttles the associated Ca2+ /calmodulin acting as an activating stimulus of the Ca2+ signal transduction [49]. 3.5. Expression and Inactivation of Cav 1.2 in the Absence of the β and α2 δ. Are the β and α2 δ subunits essential for the functional expression of the Cav 1.2 channel? The analysis of the effects of exogenous CaM (CaMex ) on the expression and properties of Cav 1.2 in the absence of either β [18] or α2 δ [57] clearly demonstrated that neither β nor α2 δ is essential. Overexpression of CaMex only slightly modifies the voltage gating of the α1C /β2d /α2 δ channel by shifting the voltage dependence of activation and inactivation towards

ISRN Molecular Biology more negative potentials, facilitating (but not accelerating) inactivation, and increasing the density of ICa approximately 2-fold [18]. CDI is retained, as it is evident from the effect of the replacement of Ca2+ for Ba2+ as the charge carrier that significantly increased the time course of inactivation of the current (Figure 5(A)). New understanding of the roles of β and α2 δ comes with the finding that CaMex renders expression and activity of the α1C channel in the absence of β (Figure 5(B)) or α2 δ (Figure 5(C)), but not both of these auxiliary subunits. Although CaMex is structurally unrelated to β and α2 δ, it supports trafficking, CDI, and channel gating. Quantitative analysis showed that CaMex did not stimulate redistribution of α1C in PM over the cytoplasm, but significantly enhanced plasma membrane targeting of α1C /α2 δ channels. On the other hand, CaMex did not enhance the relative distribution of the α1C /β2d and α1C /β2d /α2 δ channels in the plasma membrane over the cytoplasm. Thus, depending on the auxiliary subunit present, CaMex -supported channel activity of α1C /β2d and α1C /α2 δ is under control of different mechanisms. In spite of that, the CaMex -facilitated, single-auxiliary-subunit channels exhibit quite similar properties including significantly slower inactivation kinetics of the calcium current and a strong shift of the voltage dependence of activation and inactivation towards more negative potentials. Similar to the conventional α1C /β2d /α2 δ channels, these channels retain CDI and high sensitivity to dihydropyridine calcium channel blockers [18]. However, only the α1C /β/CaMex channel shows facilitation of the calcium current by strong depolarization prepulse [57] (data not shown). Because CaM associated with CBD is involved in CDI, it is clear that the effect of CaMex is mediated by different CaM-binding site(s). One of the potential candidates of such a site is present in the distal part of the α1C Nterminal tail [58, 59]. It remains to be seen whether this site indeed plays an integrating role in the regulatory bundle of several molecular determinants supporting Cav 1.2 inactivation. Another possibility confines the role of CaMex to the activation of silent Cav 1.2 within the large clusters, where limited local availability of CaM may be the reason of the low fractional activity described in Section 2.5. Whatever the mechanisms associated with regulation of Cav 1.2 by CaM are, they seems to have little practical implication for use in medicine at this time exactly because CaM is a ubiquitous and multifunctional peptide that regulates many other cellular functions, while its presence in Cav 1.2 is vital for CDI.

4. β-Subunit Modulation of Cav 1.2 Remarkable molecular variability of β subunits, reflected in altered inactivation properties of the differentially modulated Cav 1.2 [12, 60], exemplified in Figure 3(A) (panel e) presents a new opportunity for the development of innovative approaches to the treatment of the diseases associated with Ca2+ mishandling. Several recent observations provide a foundation for such an optimistic view. First, β subunits exhibit a tendency to form homo- and hetero-oligomers [11, 61] that was directly demonstrated by a variety of

ISRN Molecular Biology

7

α1C /β2d /α2 δ/CAMex (a)

(b) Ca2+

Ba2+

400 pA 200 ms

(A) α1C /α2 δ/CAMex (a)

(b) Ca2+

Ba2+

50 pA 200 ms

(B) α1C /β2d /CAMex

(b) Ca2+

(a)

Ba2+

200 pA 200 ms

(C)

Figure 5: Activity of Cav 1.2 expressed in the absence of β or α2 δ subunits. Shown are (a) epifluorescent images illustrating the predominant PM localization of EYFPN -α1C (scaling bars, 4 μm) in COS1 cells and (b) superimposed traces of the maximum IBa (black) and ICa (red) scaled to the same amplitude for α1C /β2d /α2 δ/CaMex (A), β-free α1C /α2 δ/CaMex [18] (B) and α2 δ-free α1C /β2d /CaMex channel [57] (C).

biochemical techniques in both native cells and in recombinant expression system. While an augmentation of β homooligomerization significantly increases the density of ICa , heterooligomerization of β2 splice variants with other β subunits may also change the voltage-dependence and inactivation kinetics of Cav 1.2 [11]. The β-oligomerization is mediated by several molecular determinants and thus needs multiple interventions to be managed, for example, in case of pathogenic overexpression of β2 . However, it seems to be more feasible to target β2 itself; molecular determinant of β2 -specific slow and incomplete inactivation (see Figure 2(A), panel d) was identified [46] as the 40amino-acid C-terminal determinant (β2 CED) present in all 7 known naturally occurring β2 splice variants. Uncoupling of its Ca2+ - and CaM-independent interaction with CBD (Figure 3(A), panel a) recovers the inactivation properties characteristic for β1b /β3 -modulated Cav 1.2 exhibiting rapid and complete inactivation of ICa , as it was shown in deletion experiments. In my view, such selective uncoupling of β2 CED from binding to its receptor in CBD is a new attractive strategy to manage Ca2+ overload because other β subunits are not to be affected. Moreover, a cross-talk between Cav 1.2 and the nearest target Ca2+ /CaM-dependent protein kinase II [62, 63] will be preserved.

5. Conclusions This paper has demonstrated that we know how to accelerate inactivation of Cav 1.2 to τ f less than 10 ms (Figure 3(C)),

to deprive it from inactivation completely (Figures 4(B) and 4(C)), or to eliminate dependence of its expression from β or α2 δ without significant consequences for inactivation. We outlined the ultimate roles of the α1C termini and CaM for inactivation, and yet none of these studies have brought us any closer to the ultimate goal of managing calcium mishandling associated with Cav 1.2 except of old and, unfortunately, not too selective calcium channel blockers. The only new feasible target is pathogenic β2 modulation of Cav 1.2, where effector-receptor interaction is established. In terms of molecular biology, Cav 1.2 is certainly among the most complicated regulatory systems known. Remarkable molecular diversity of each of the Cav 1.2 constituents gives rise to multiple genetic/splice variants of the channel that are subject to segregation into large and diverse clusters and to continuous functional change through homo- and hetero-oligomerization of β and other signaling components, not to speak about species, tissue, and developmental variability. We are surprised by the redundancy of the properties of multiple Cav 1.2 isoforms [64, 65] and are even more surprised when some of them, showing just “conventional” electrophysiological properties, turn out to be associated with a disease [13]. In looking for an explanation, our insight should not be intuitively focused just on the characteristics of the calcium current-voltage dependence, amplitude, and duration. The end response, such as spatial and temporal organization of CREB signaling events associated with specific Cav 1.2 isoform [66], and its competition with other (e.g., cAMP dependent) signaling mechanisms, or other Cav 1.2 isoforms present, may provide new ideas and open new frontiers for investigation of the roles of individual Cav 1.2 splice variants in normal and diseased cells and tissues.

Abbreviations ADSI: CaM: CBD: CDI: ECFP: EYFP: FALI: FI: FRET: GFP: SI:

Annual determinant of slow inactivation Calmodulin Calmodulin-binding domain Ca2+ -dependent inactivation Enhanced cyan fluorescent protein Enhanced yellow fluorescent protein Fluorophore-assisted light inactivation Fast inactivation Fluorescent resonance energy transfer Green fluorescent protein Slow voltage-dependent inactivation.

References [1] D. M. Bers, “Calcium cycling and signaling in cardiac myocytes,” Annual Review of Physiology, vol. 70, no. 1, pp. 23– 49, 2008. [2] M. T. Nelson, N. B. Standen, J. E. Brayden, and J. F. Worley, “Noradrenaline contracts arteries by activating voltagedependent calcium channels,” Nature, vol. 336, no. 6197, pp. 382–385, 1988. [3] H. Bading, D. D. Ginty, and M. E. Greenberg, “Regulation of gene expression in hippocampal neurons by distinct calcium

8

[4]

[5]

[6] [7]

[8]

[9]

[10]

[11]

[12]

[13]

[14] [15]

[16]

[17]

[18]

[19]

ISRN Molecular Biology signaling pathways,” Science, vol. 260, no. 5105, pp. 181–186, 1993. D. Johnston, S. Williams, D. Jaffe, and R. Gray, “NMDAreceptor-independent long-term potentiation,” Annual Review of Physiology, vol. 54, pp. 489–505, 1992. C. R. Artalejo, M. E. Adams, and A. P. Fox, “Three types of Ca2+ channel trigger secretion with different efficacies in chromaffin cells,” Nature, vol. 367, no. 6458, pp. 72–76, 1994. E. Carafoli, “Intracellular calcium homeostasis.,” Annual Review of Biochemistry, vol. 56, pp. 395–433, 1987. V. Flockerzi, H. J. Oeken, and F. Hofmann, “Purification of a functional receptor for calcium-channel blockers from rabbit skeletal-muscle microsomes,” European Journal of Biochemistry, vol. 161, no. 1, pp. 217–224, 1986. M. Takahashi, M. J. Seagar, J. F. Jones, B. F. X. Reber, and W. A. Catterall, “Subunit structure of dihydropyridine-sensitive calcium channels from skeletal muscle,” Proceedings of the National Academy of Sciences of the United States of America, vol. 84, no. 15, pp. 5478–5482, 1987. A. C. Dolphin, “β Subunits of voltage-gated calcium channels,” Journal of Bioenergetics and Biomembranes, vol. 35, no. 6, pp. 599–620, 2003. N. Klugbauer, E. Marais, and F. Hofmann, “Calcium channel α2 δ subunits: differential expression, function, and drug binding,” Journal of Bioenergetics and Biomembranes, vol. 35, no. 6, pp. 639–647, 2003. Q. Z. Lao, E. Kobrinsky, Z. Liu, and N. M. Soldatov, “Oligomerization of Cav β subunits is an essential correlate of Ca2+ channel activity,” FASEB Journal, vol. 24, no. 12, pp. 5013–5023, 2010. H. M. Colecraft, B. Alseikhan, S. X. Takahashi et al., “Novel functional properties of Ca2+ channel β subunits revealed by their expression in adult rat heart cells,” Journal of Physiology, vol. 541, no. 2, pp. 435–452, 2002. S. Tiwari, Y. Zhang, J. Heller, D. R. Abernethy, and N. M. Soldatov, “Artherosclerosis-related molecular alteration of the human Cav 1.2 calcium channel α1C subunit,” Proceedings of the National Academy of Sciences of the United States of America, vol. 103, no. 45, pp. 17024–17029, 2006. N. M. Soldatov, “Genomic structure of human L-type Ca2+ channel,” Genomics, vol. 22, no. 1, pp. 77–87, 1994. S. Berjukow, F. D¨oring, M. Froschmayr, M. Grabner, H. Glossmann, and S. Hering, “Endogenous calcium channels in human embryonic kidney (HEK293) cells,” British Journal of Pharmacology, vol. 118, no. 3, pp. 748–754, 1996. M. Kurejov´a, B. Uhr´ık, Z. Sulov´a, B. Sedl´akov´a, O. Kriˇzanov´a, and L. Lacinov´a, “Changes in ultrastructure and endogenous ionic channels activity during culture of HEK 293 cell line,” European Journal of Pharmacology, vol. 567, no. 1-2, pp. 10– 18, 2007. A. Meir, D. C. Bell, G. J. Stephens, K. M. Page, and A. C. Dolphin, “Calcium channel β subunit promotes voltagedependent modulation of α1B by Gβγ,” Biophysical Journal, vol. 79, no. 2, pp. 731–746, 2000. A. Ravindran, Q. Z. Lao, J. B. Harry, P. Abrahimi, E. Kobrinsky, and N. M. Soldatov, “Calmodulin-dependent gating of Cav 1.2 calcium channels in the absence of Cav β subunits,” Proceedings of the National Academy of Sciences of the United States of America, vol. 105, no. 23, pp. 8154–8159, 2008. E. Kobrinsky, K. J. F. Kepplinger, A. Yu et al., “Voltage-gated rearrangements associated with differential β-subunit modulation of the L-type Ca2+ channel inactivation,” Biophysical Journal, vol. 87, no. 2, pp. 844–857, 2004.

[20] E. Kobrinsky, J.-H. Lee, and N. M. Soldatov, “Selective fluorophore-assisted light inactivation of voltage-gated calcium channels,” Channels, vol. 6, no. 3, pp. 154–156, 2012. [21] E. Kobrinsky, E. Schwartz, D. R. Abernethy, and N. M. Soldatov, “Voltage-gated mobility of the Ca2+ channel cytoplasmic tails and its regulatory role,” Journal of Biological Chemistry, vol. 278, no. 7, pp. 5021–5028, 2003. [22] S. Herzig, I. F. Y. Khan, D. Gr¨undemann et al., “Mechanism of Cav 1.2 channel modulation by the amino terminus of cardiac β2-subunits,” FASEB Journal, vol. 21, no. 7, pp. 1527–1538, 2007. [23] W. Y. W. Lew, L. V. Hryshko, and D. M. Bers, “Dihydropyridine receptors are primarily functional L-type calcium channels in rabbit ventricular myocytes,” Circulation Research, vol. 69, no. 4, pp. 1139–1145, 1991. [24] J. W. Hell, R. E. Westenbroek, C. Warner et al., “Identification and differential subcellular localization of the neuronal class C and class D L-type calcium channel α1 subunits,” Journal of Cell Biology, vol. 123, no. 4, pp. 949–962, 1993. [25] D. Lipscombe, D. V. Madison, M. Poenie, H. Reuter, R. Y. Tsien, and R. W. Tsien, “Spatial distribution of calcium channels and cytosolic calcium transients in growth cones and cell bodies of sympathetic neurons,” Proceedings of the National Academy of Sciences of the United States of America, vol. 85, no. 7, pp. 2398–2402, 1988. [26] R. E. Westenbroek, M. K. Ahlijanian, and W. A. Catterall, “Clustering of L-type Ca2+ channels at the base of major dendrites in hippocampal pyramidal neurons,” Nature, vol. 347, no. 6290, pp. 281–284, 1990. [27] C. Franzini-Armstrong, F. Protasi, and P. Tijskens, “The assembly of calcium release units in cardiac muscle,” Annals of the New York Academy of Sciences, vol. 1047, pp. 76–85, 2005. [28] D. V. Gathercole, D. J. Colling, J. N. Skepper, Y. Takagishi, A. J. Levi, and N. J. Severs, “Immunogold-labeled Ltype calcium channels are clustered in the surface plasma membrane overlying junctional sarcoplasmic reticulum in guinea-pig myocytes—implications for excitation-contraction coupling in cardiac muscle,” Journal of Molecular and Cellular Cardiology, vol. 32, no. 11, pp. 1981–1994, 2000. [29] Y. Takagishi, K. Yasui, N. J. Severs, and Y. Murata, “Speciesspecific difference in distribution of voltage-gated L-type Ca2+ channels of cardiac myocytes,” American Journal of Physiology, vol. 279, no. 6, pp. C1963–C1969, 2000. [30] G. S. Harms, L. Cognet, P. H. M. Lommerse et al., “Singlemolecule imaging of L-type Ca2+ channels in live cells,” Biophysical Journal, vol. 81, no. 5, pp. 2639–2646, 2001. [31] E. Kobrinsky, P. Abrahimi, S. Q. Duong et al., “Effect of Cav β subunits on structural organization of Cav 1.2 calcium channels,” PLoS ONE, vol. 4, no. 5, Article ID e5587, 2009. [32] K. S. Lee, E. Marban, and R. W. Tsien, “Inactivation of calcium channels in mammalian heart cells: joint dependence on membrane potential and intracellular calcium,” Journal of Physiology, vol. 364, pp. 395–411, 1985. [33] X. Zong and F. Hofmann, “Ca2+ -dependent inactivation of the class C L-type Ca2+ channel is a property of the α1 subunit,” FEBS Letters, vol. 378, no. 2, pp. 121–125, 1996. [34] N. M. Soldatov, R. D. Z¨uhlke, A. Bouron, and H. Reuter, “Molecular structures involved in L-type calcium channel inactivation. Role of the carboxyl-terminal region encoded by exons 40–42 in α1C subunit in the kinetics and Ca2+ dependence of inactivation,” Journal of Biological Chemistry, vol. 272, no. 6, pp. 3560–3566, 1997.

ISRN Molecular Biology [35] N. M. Soldatov, M. Oz, K. A. O’Brien, D. R. Abernethy, and M. Morad, “Molecular determinants of L-type Ca2+ channel inactivation: segment exchange analysis of the carboxylterminal cytoplasmic motif encoded by exons 40-42 of the human α1C subunit gene,” Journal of Biological Chemistry, vol. 273, no. 2, pp. 957–963, 1998. [36] R. D. Z¨uhlke and H. Reuter, “Ca2+ -sensitive inactivation of Ltype Ca2+ channels depends on multiple cytoplasmic amino acid sequences of the α1C subunit,” Proceedings of the National Academy of Sciences of the United States of America, vol. 95, no. 6, pp. 3287–3294, 1998. [37] B. Z. Peterson, C. D. DeMaria, J. P. Adelman, and D. T. Yue, “Calmodulin is the Ca2+ sensor for Ca2+ -dependent inactivation of L-type calcium channels,” Neuron, vol. 22, no. 3, pp. 549–558, 1999. [38] N. Qin, R. Olcese, M. Bransby, T. Lin, and L. Birnbaumer, “Ca2+ -induced inhibition of the cardiac Ca2+ channel depends on calmodulin,” Proceedings of the National Academy of Sciences of the United States of America, vol. 96, no. 5, pp. 2435– 2438, 1999. [39] R. D. Z¨uhlke, G. S. Pitt, K. Deisseroth, R. W. Tsien, and H. Reuter, “Calmodulin supports both inactivation and facilitation of L-type calcium channels,” Nature, vol. 399, no. 6732, pp. 159–162, 1999. [40] M. G. Erickson, B. A. Alseikhan, B. Z. Peterson, and D. T. Yue, “Preassociation of calmodulin with voltage-gated Ca2+ channels revealed by FRET in single living cells,” Neuron, vol. 31, no. 6, pp. 973–985, 2001. [41] P. Pate, J. Mochca-Morales, Y. Wu et al., “Determinants for calmodulin binding on voltage-dependent Ca2+ channels,” Journal of Biological Chemistry, vol. 275, no. 50, pp. 39786– 39792, 2000. [42] C. Romanin, R. Gamsjaeger, H. Kahr et al., “Ca2+ sensors of Ltype Ca2+ channel,” FEBS Letters, vol. 487, no. 2, pp. 301–306, 2000. [43] M. X. Mori, M. G. Erickson, and D. T. Yue, “Functional stoichiometry and local enrichment of calmodulin interacting with Ca2+ channels,” Science, vol. 304, no. 5669, pp. 432–435, 2004. [44] L. Xiong, Q. K. Kleerekoper, R. He, J. A. Putkey, and S. L. Hamilton, “Sites on calmodulin that interact with the C-terminal tail of Cav 1.2 channel,” Journal of Biological Chemistry, vol. 280, no. 8, pp. 7070–7079, 2005. [45] R. Zhang, I. Dzhura, C. E. Grueter, W. Thiel, R. J. Colbran, and M. E. Anderson, “A dynamic α-β inter-subunit agonist signaling complex is a novel feedback mechanism for regulating Ltype Ca2+ channel opening,” FASEB Journal, vol. 19, no. 11, pp. 1573–1575, 2005. [46] Q. Z. Lao, E. Kobrinsky, J. B. Harry, A. Ravindran, and N. M. Soldatov, “New determinant for the Cav β2 subunit modulation of the Cav 1.2 calcium channel,” Journal of Biological Chemistry, vol. 283, no. 23, pp. 15577–15588, 2008. [47] E. Kobrinsky, S. Tiwari, V. A. Maltsev et al., “Differential role of the α1C subunit tails in regulation of the Cav 1.2 channel by membrane potential, β subunits, and Ca2+ ions,” Journal of Biological Chemistry, vol. 280, no. 13, pp. 12474–12485, 2005. [48] S. Hering, S. Berjukow, S. Acz´el, and E. N. Timin, “Ca2+ channel block and inactivation: common molecular determinants,” Trends in Pharmacological Sciences, vol. 19, no. 11, pp. 439– 443, 1998. [49] N. M. Soldatov, “Ca2+ channel moving tail: link between Ca2+ induced inactivation and Ca2+ signal transduction,” Trends in Pharmacological Sciences, vol. 24, no. 4, pp. 167–171, 2003.

9 [50] S. Hering, S. Berjukow, S. Sokolov et al., “Molecular determinants of inactivation in voltage-gated Ca2+ channels,” Journal of Physiology, vol. 528, no. 2, pp. 237–249, 2000. [51] N. M. Soldatov, S. Zhenochin, B. AlBanna, D. R. Abernethy, and M. Morad, “New molecular determinant for inactivation of the human L-type α1C Ca2+ channel,” Journal of Membrane Biology, vol. 177, no. 2, pp. 129–135, 2000. [52] J. F. Zhang, P. T. Ellinor, R. W. Aldrich, and R. W. Tsien, “Molecular determinants of voltage-dependent inactivation in calcium channels,” Nature, vol. 372, no. 6501, pp. 97–100, 1994. [53] C. Shi and N. M. Soldatov, “Molecular determinants of voltage-dependent slow inactivation of the Ca2+ channel,” Journal of Biological Chemistry, vol. 277, no. 9, pp. 6813–6821, 2002. [54] M. Morad and N. Soldatov, “Calcium channel inactivation: possible role in signal transduction and Ca2+ signaling,” Cell Calcium, vol. 38, no. 3-4, pp. 223–231, 2005. [55] Y. Blumenstein, N. Kanevsky, G. Sahar, R. Barzilai, T. Ivanina, and N. Dascal, “A novel long N-terminal isoform of human Ltype Ca2+ channel is up-regulated by protein kinase C,” Journal of Biological Chemistry, vol. 277, no. 5, pp. 3419–3423, 2002. [56] E. Shistik, T. Ivanina, Y. Blumenstein, and N. Dascal, “Crucial role of N terminus in function of cardiac L-type Ca2+ channel and its modulation by protein kinase C,” Journal of Biological Chemistry, vol. 273, no. 28, pp. 17901–17909, 1998. [57] A. Ravindran, E. Kobrinsky, Q. Z. Lao, and N. M. Soldatov, “Functional properties of the Cav 1.2 calcium channel activated by calmodulin in the absence of α2 δ subunits,” Channels, vol. 3, no. 1, pp. 25–31, 2009. [58] A. Benmocha, L. Almagor, S. Oz, J. A. Hirsch, and N. Dascal, “Characterization of the calmodulin-binding site in the N terminus of Cav 1.2 .,” Channels, vol. 3, no. 5, pp. 337–342, 2009. [59] T. Ivanina, Y. Blumenstein, E. Shistik, R. Barzilai, and N. Dascal, “Modulation of L-type Ca2+ channels by Gβγ and calmodulin via interactions with N and C termini of α1C ,” Journal of Biological Chemistry, vol. 275, no. 51, pp. 39846– 39854, 2000. [60] Z. Buraei and J. Yang, “The β subunit of voltage-gated Ca2+ channels,” Physiological Reviews, vol. 90, no. 4, pp. 1461–1506, 2010. [61] E. Miranda-Laferte, G. Gonzalez-Gutierrez, S. Schmidt et al., “Homodimerization of the Src homology 3 domain of the calcium channel β-subunit drives dynamin-dependent endocytosis,” Journal of Biological Chemistry, vol. 286, no. 25, pp. 22203–22210, 2011. [62] C. E. Grueter, S. A. Abiria, Y. Wu, M. E. Anderson, and R. J. Colbran, “Differential regulated interactions of calcium/calmodulin-dependent protein kinase II with isoforms of voltage-gated calcium channel β subunits,” Biochemistry, vol. 47, no. 6, pp. 1760–1767, 2008. [63] O. M. Koval, X. Guan, Y. Wu et al., “Cav 1.2 β-subunit coordinates CaMKII-triggered cardiomyocyte death and afterdepolarizations,” Proceedings of the National Academy of Sciences of the United States of America, vol. 107, no. 11, pp. 4996–5000, 2010. [64] P. Liao, D. Yu, S. Lu et al., “Smooth muscle-selective alternatively spliced exon generates functional variation in Cav 1.2 calcium channels,” Journal of Biological Chemistry, vol. 279, no. 48, pp. 50329–50335, 2004.

10 [65] P. Liao, D. Yu, G. Li et al., “A smooth muscle Cav 1.2 calcium channel splice variant underlies hyperpolarized window current and enhanced state-dependent inhibition by nifedipine,” Journal of Biological Chemistry, vol. 282, no. 48, pp. 35133– 35142, 2007. [66] E. Kobrinsky, S. Q. Duong, A. Sheydina, and N. M. Soldatov, “Microdomain organization and frequency-dependence of CREB-dependent transcriptional signaling in heart cells,” FASEB Journal, vol. 25, no. 5, pp. 1544–1555, 2011.

ISRN Molecular Biology